Action tremor features discovery for essential tremor and Parkinson's disease with explainable multilayer BiLSTM

被引：0
作者：
Teo, Yu Xuan [1 ]
Lee, Rui En [1 ]
Nurzaman, Surya Girinatha [2 ]
Tan, Chee Pin [1 ]
Chan, Ping Yi [1 ]
机构：
[1] Department of Electrical & Robotics Engineering, School of Engineering, Monash University Malaysia
[2] Department of Mechanical Engineering, School of Engineering, Monash University Malaysia, Bandar Sunway
来源：
Computers in Biology and Medicine | 2024年 / 180卷
关键词：
Classification; Deep learning; Essential tremor; Explainable AI; Parkinson's tremor;
D O I：
10.1016/j.compbiomed.2024.108957
中图分类号：
学科分类号：
摘要：
The tremors of Parkinson's disease (PD) and essential tremor (ET) are known to have overlapping characteristics that make it complicated for clinicians to distinguish them. While deep learning is robust in detecting features unnoticeable to humans, an opaque trained model is impractical in clinical scenarios as coincidental correlations in the training data may be used by the model to make classifications, which may result in misdiagnosis. This work aims to overcome the aforementioned challenge of deep learning models by introducing a multilayer BiLSTM network with explainable AI (XAI) that can better explain tremulous characteristics and quantify the respective discovered important regions in tremor differentiation. The proposed network classifies PD, ET, and normal tremors during drinking actions and derives the contribution from tremor characteristics, (i.e., time, frequency, amplitude, and actions) utilized in the classification task. The analysis shows that the XAI-BiLSTM marks the regions with high tremor amplitude as important in classification, which is verified by a high correlation between relevance distribution and tremor displacement amplitude. The XAI-BiLSTM discovered that the transition phases from arm resting to lifting (during the drinking cycle) is the most important action to classify tremors. Additionally, the XAI-BiLSTM reveals frequency ranges that only contribute to the classification of one tremor class, which may be the potential distinctive feature to overcome the overlapping frequencies problem. By revealing critical timing and frequency patterns unique to PD and ET tremors, this proposed XAI-BiLSTM model enables clinicians to make more informed classifications, potentially reducing misclassification rates and improving treatment outcomes. © 2024 The Authors
引用
收藏
相关论文
共 57 条
  • [1] Hossen A., Deuschl G., Groppa S., Heute U., Muthuraman M., Discrimination of physiological tremor from pathological tremor using accelerometer and surface EMG signals, Technol. Health Care, 28, 5, pp. 461-476, (2020)
  • [2] Thenganatt M.A., Louis E.D., Distinguishing essential tremor from Parkinson's disease: bedside tests and laboratory evaluations, Expert Rev. Neurother., 12, 6, pp. 687-696, (2012)
  • [3] Kraus P.H., Lemke M.R., Reichmann H., Kinetic tremor in Parkinson's disease – an underrated symptom, J. Neural. Transm., 113, 7, pp. 845-853, (2006)
  • [4] Tolosa E., Wenning G., Poewe W., The diagnosis of Parkinson's disease, Lancet Neurol., 5, 1, pp. 75-86, (2006)
  • [5] Lilhore U.K., Et al., Hybrid CNN-LSTM model with efficient hyperparameter tuning for prediction of Parkinson's disease, Sci. Rep., 13, 1, (2023)
  • [6] Golub T.R., Et al., Molecular classification of cancer: class discovery and class prediction by gene expression monitoring, Science, 286, 5439, pp. 531-537, (1999)
  • [7] Pastorino M., Arredondo M.T., Cancela J., Guillen S., Wearable sensor network for health monitoring: the case of Parkinson disease, J. Phys. Conf., 450, 1, (2013)
  • [8] Hong J., Et al., Convolutional neural network-based annotation of bacterial type IV secretion system effectors with enhanced accuracy and reduced false discovery, Briefings Bioinf., 21, (2019)
  • [9] Wang Y., Et al., A task-specific encoding algorithm for RNAs and RNA-associated interactions based on convolutional autoencoder, Nucleic Acids Res., 51, 21, (2023)
  • [10] Hong J., Et al., Protein functional annotation of simultaneously improved stability, accuracy and false discovery rate achieved by a sequence-based deep learning, Briefings Bioinf., 21, 4, pp. 1437-1447, (2020)
123456